Journal of Animal Ecology 2014, 83, 1025–1034
doi: 10.1111/1365-2656.12208
Ecological opportunities and intraspecific competition
alter trophic niche specialization in an opportunistic
stream predator
Charlotte Evangelista1,2, Anatole Boiche3,4, Antoine Lecerf3,4 and Julien Cucherousset1,2*
CNRS, Université Paul Sabatier, ENFA, UMR5174 EDB (Laboratoire Evolution
& Diversite Biologique), 118 route de
Narbonne, F-31062 Toulouse, France; 2Universite Toulouse 3 Paul Sabatier, CNRS, UMR 5174 EDB, F-31062
Toulouse, France; 3UPS, INP, EcoLab (Laboratoire Ecologie Fonctionnelle et Environnement), Universite de
Toulouse, 118 route de Narbonne, F-31062 Toulouse, France; and 4CNRS, EcoLab, F-31062 Toulouse, France
1
Summary
1. Many generalist populations are composed of specialized individuals that use a narrow
part of the population’s niche. Ecological theories predict that individual specialization and
population trophic niche are determined by biotic interactions and resource diversity emerging from environmental variations (i.e. ecological opportunities). However, due to the paucity
of empirical and experimental demonstrations, the genuine importance of each of these drivers in determining trophic niche attributes is not fully appreciated.
2. The present study aimed at determining the population level and individual responses of
brown trout (Salmo trutta) to variations in ecological opportunities (terrestrial prey inputs)
and autochthonous prey communities among 10 stream reaches along a riparian condition
gradient using individual longitudinal monitoring and stable isotope analyses.
3. Our results suggested that trophic niche diversity varied along the environmental gradient,
while individual trophic specialization was indirectly driven by ecological opportunities
through strengthened intraspecific competition. Individual diet was repeatable over the study
period, and the growth rate of juvenile brown trout increased with their specialization for
aquatic predatory invertebrates.
4. Our findings highlight the dual influences of intraspecific competition and ecological
opportunities on individual trophic specialization and population trophic niche.
Key-words: competition, individual performances, inter-individual variability, riparian landuse, stable isotope analyses, trophic subsidies
Introduction
Ecologists have long considered populations as a whole,
neglecting traits variation among conspecific individuals
and its significant ecological and evolutionary consequences (Bolnick et al. 2003). Morphological variations
among individuals within populations are frequently
related to niche differentiation through the use of different habitats and trophic resources to optimize foraging
efficiency and subsequently increase individual performances (Bolnick et al. 2003; Svanb€ack & Bolnick 2008;
Cucherousset et al. 2011). However, in some cases, individuals within a population can exhibit different trophic
niches and display slight (if no) morphological differences
*Correspondence author. E-mail: [email protected]
(Sk
ulason & Smith 1995; Tinker, Bentall & Estes 2008).
This ‘individual trophic specialization’ is a widespread
phenomenon (Bolnick et al. 2003), and a more comprehensive understanding of the mechanisms driving individual specialization is required (Ara
ujo, Bolnick &
Layman 2011).
Individual trophic specialization in wild populations
has been mainly demonstrated to emerge from biotic
interactions such as competition and predation (Svanb€
ack
& Persson 2004; Ekl€
ov & Svanb€
ack 2006; Bolnick et al.
2010). However, the relative contributions of competition
and predation to individual trophic specialization are
highly context dependent, and the degree of individual
specialization may also depend on the diversity and/or
availability of food resources. Ecological opportunities,
that is, the resource diversity emerging from environmental variations (Ara
ujo, Bolnick & Layman 2011), can
© 2014 The Authors. Journal of Animal Ecology © 2014 British Ecological Society
1026 C. Evangelista et al.
strongly modify the occurrence and/or intensity of trophic
specialization (Quevedo, Svanb€ack & Ekl€
ov 2009) but few
studies have investigated this phenomenon in wild populations (e.g. Ara
ujo, Bolnick & Layman 2011).
Trophic subsidies (i.e. spatial flows of food resources
across ecosystem boundaries), by enhancing the availability, diversity, and/or temporal fluctuations of food
resources, have the potential to promote individual trophic specialization (Ara
ujo, Bolnick & Layman 2011).
Optimal utilization of trophic subsidies often involves
changes in foraging behaviour and habitat use of consumers in response to the spatial segregation of allochthonous
and autochthonous resources. Studies on pulsed marinederived resources, through the anadromous migration of
spawning salmons, have shown that terrestrial predators
foraging on these trophic subsidies are more specialized
than individuals that are spatially restricted to forage
uniquely on terrestrial prey (Darimont, Paquet & Reimchen 2009; Wipfli & Baxter 2010).
Trophic subsidies have been reported to mediate human
impacts on ecosystems. For instance, commercial fisheries
and habitat alteration have led to drastic declines in
migratory salmonids stocks (Allendorf & Hard 2009) with
subsequent consequences on the fluxes of marine-derived
nutrients in freshwater ecosystems (Schindler et al. 2005).
Invasive species can also modify the transfer of allochthonous energy in recipient ecosystem (Baxter et al. 2004),
with some invasive individuals displaying trophic specialization for allochthonous prey (Cucherousset et al. 2012).
Across the stream–forest interface, allochthonous
inputs, principally composed of nutrients, plant detritus
and organisms are used by a wide range of consumers
inhabiting the recipient ecosystem (Baxter, Fausch &
Saunders 2005; Bartels et al. 2012). Previous studies have
highlighted the importance of terrestrial invertebrates to
the diet and annual energy budget of many stream fish
(e.g. Kawaguchi & Nakano 2001; Baxter et al. 2007; Bartels et al. 2012). For instance, substantial inputs of terrestrial invertebrates from a productive forest ecosystem to a
nutrient-poor stream can represent more than half of the
energy requirement of stream fish in summer (Kawaguchi
& Nakano 2001).
Riparian land-use and forest management alter the
inputs of terrestrial invertebrate prey to streams (Nakano,
Fausch & Kitano 1999a; Kawaguchi & Nakano 2001)
and community of aquatic invertebrate prey (Kiffney,
Richardson & Bull 2003; Lecerf et al. 2012). Terrestrial
plant litter is the primary energy source to forested
streams, and thus, loss of forest cover is expected to
reduce the production of aquatic invertebrate detritivores
(Wallace et al. 1997). Conversely, the production of aquatic invertebrate herbivores should increase with riparian
canopy openness as in-stream primary producers (e.g.
algae and moss) are released from light limitation (Kiffney, Richardson & Bull 2003). Riparian canopy cover also
regulates stream temperatures (Moore, Spittlehouse &
Story 2005) and, subsequently, the metabolism of aquatic
organisms. As opportunistic predators, stream salmonids
can shift their diet in response to quantitative and qualitative changes in terrestrial prey inputs and aquatic prey community to meet their nutritional requirements (Nakano,
Miyasaka & Kuhara 1999b; Baxter et al. 2004), with potential modifications of population patterns of trophic specialization and individual performances. However, despite
compelling evidence that salmonids are affected by forest
management, little is known on the genuine importance of
trophic factors in mediating individual- and populationlevel responses of fish to changes in riparian canopy cover
(Mellina & Hinch 2009; Lecerf et al. 2012).
In the present study, we investigated the trophic ecology of brown trout (Salmo trutta), an opportunistic predator, in low-order streams flowing through a deciduous
hardwood forest, using stable isotope analyses (Layman
et al. 2012). Ten stream sites were selected along a gradient of riparian canopy cover in a forested area managed
under traditional silvicultural systems, anticipating
changes in the availability and the composition of aquatic
and terrestrial prey along this environmental gradient. We
quantified trophic specialization and trophic diversity of
brown trout in each site. In addition, we determined individual diet and growth rate as a fitness proxy to infer possible ecological benefits of individual specialization
(Bolnick 2004). We first predicted that terrestrial prey
inputs are ecological opportunities for brown trout and
increased under dense canopies. Then, we predicted that
dense riparian canopy cover should promote individual
trophic specialization and influence population trophic
niche attributes in brown trout. In addition to the ecological opportunities hypothesis, we also evaluated the importance of intraspecific competition in mediating the brown
trout response to variation in riparian canopy cover.
Materials and methods
study area and site selection
The present study was conducted from April to September 2011
in the Montagne Noire (South-Western France; 43°330 N, 1°290 E),
a low altitude mountain region drained by a high density of
headwater streams. Catchments are historically covered by European beech (Fagus sylvatica) forest. During the twentieth century,
upland forests were converted into commercial beech production
and conifer plantations. Valley bottoms and riparian forest are
still managed under traditional silvicultural systems. For instance,
several areas comprise oak coppice-with-beech standards, where
natural regenerating vegetation is preserved. A few areas are
affected by riparian clear-cut logging. However, due to the small
size of harvesting patches (< 5% of total catchment area) and the
protection measures taken to minimize disturbance by logging
machines, the ecological impact on stream ecosystems, other than
those mediated by changes in riparian canopy structure and composition, was small (Lecerf et al. 2012).
Brown trout was the only fish species present in the study
streams. To assess the effect of changes in forest canopy cover
on its trophic ecology, we selected 10 stream sites representing a
© 2014 The Authors. Journal of Animal Ecology © 2014 British Ecological Society, Journal of Animal Ecology, 83, 1025–1034
Individual variability in disturbed streams
gradient of riparian canopy openness, an integrative metric to
reveal the importance of riparian vegetation on aquatic ecosystem
properties (Lecerf et al. 2012). Sites were situated on the lower
third section of c. 500-m-long stream reaches running through
homogeneous riparian vegetation in terms of canopy structure
and species composition. Site length ranged from 554 to 960 m
and was also selected to reduce potential confounding effects of
stream habitat features on brown trout (Appendix S1, Supporting
information). All sites were located on different streams with the
exception of two sites that were considered, however, as independent observations because they were located 13 km apart. Mean
canopy openness was quantified as canopy gap area within a
0–60° zenith range based on four digital hemispherical pictures
taken from the streambed in early summer 2011 at fully foliated
stage (Lecerf et al. 2012). Image analyses were performed using
the software ‘Gap Light Analyzer v 2.0’ (Simon Fraser University, Canada). Differences in riparian forest canopy openness
were due to variable forest management intensity (coppicing or
partial harvesting), forest age and natural determinants of canopy
structure (i.e. forest blowdown, soil properties and topography).
The most open sites were selected in young (5–10 year old) forests regenerating from clear-cutting. Canopy gaps represented
< 1% of sky surface at the least open site and reached nearly
50% at the most open site. The remaining eight stream sites were
rather evenly distributed over the range of canopy openness
despite a gap in the 86–186% range (Table S1, Supporting
information). Canopy openness was significantly different
between sites (ANOVA, F9,30 = 39, P < 0001), and within-site variability measured on four pictures was low (residual variance
accounted for 8% of total variance).
habitat characteristics and trophic
resources availability
Water temperature was recorded automatically from March 2011
to September 2011 every two hours using HOBO Pendantâ Temperature loggers (Onset company, Cape Cod, MA, USA)
anchored to an iron stick placed in the downstream part of each
study site. Channel width was determined using transects set
every 10 m along each site in spring 2012. Water depth was
determined at three equally spaced points along each transect.
The proportion of riffle and pool habitats occurring between
transects was visually determined to estimate per cent wet area
covered by each habitat.
Aquatic prey availability was quantified by estimating benthic
invertebrate biomass on the 21 June 2011–23 June 2011, that is,
during the period of fish monitoring (see details below) and of
maximal benthic invertebrate abundance. Surber samples
(25 9 25 cm; 250-lm mesh size) were randomly taken in riffle
(n = 4) and pool (n = 4) habitats in each site. The flux of terrestrial prey falling into the streams was estimated using three plastic pan-traps (55 9 37 cm; 15 cm deep) deployed along the
stream banks of each site on the 1 August 2011, when terrestrial
invertebrate falling into stream peaked and weather conditions
were stable. This sampling effort (mean = 004 pan-trap per
metre) falls within a range of sampling effort reported in the literature (Kawaguchi & Nakano 2001; Nakano & Murakami 2001).
Pan-traps were filled with filtered water, and 2–3 drops of surfactant were added. After 48 h of deployment, water in pan-traps
was sieved at 250 lm to collect invertebrates. Aquatic and terrestrial invertebrate samples were preserved in 70% ethanol and
1027
stored at 15 °C until processing. In the laboratory, invertebrates
were counted and sorted to the lowest practical level (mostly genera
or family) under a binocular microscope (magnification 9 80).
Aquatic invertebrates were assigned to the trophic groups of aquatic predators, aquatic herbivores or aquatic detritivores using literature information on feeding habits and food preference (Tachet
et al. 2010). Adult aquatic insects were removed from pan-trap
samples, and remaining invertebrates were assigned to the trophic
groups of terrestrial predators or terrestrial herbivores. The habitat-weighted biomass of aquatic prey per square metre was then
calculated based on the number of individuals of each trophic
group in riffle and pool habitats, the average individual dry mass
and the proportion of each habitat type. Larvae, nymph and
imagos of each taxon had distinct weights and were considered
separately. After identification, terrestrial prey were oven-dried at
60 °C for 48 h and weighed to the nearest 001 mg for total dry
mass. The daily flux of terrestrial prey per trophic group was calculated using dry mass, pan-trap surface and trapping duration.
fish monitoring and samples analyses
Fish were initially captured and individually tagged from the 18th
to 20th of April 2011 using two-path electric fishing (EFKO FEG
1500, Leutkirch im Allgäu, Germany) in study sites delineated with
two 8-mm-mesh nets prior to sampling. Fish were anesthetized,
measured for fork length (FLC 1 mm) and weighed
(WC 01 g). For each individual, we collected scales for age
determination and a tissue sample from the left pelvic fin, which
was stored on ice and frozen for stable isotope analyses (Appendix
S2, Supporting information). Stomach contents of individuals with
FLC > 65 mm were collected by stomach flushing using a pumping
method (e.g. Nakano, Miyasaka & Kuhara 1999b) and stored in
70% ethanol. Individuals with FLC > 50 mm were tagged using
fluorescent visible implant elastomer (VIE; Northwest Marine
Technology, Shaw Island, WA, USA). Because the coding was
based on a limited number of colour combinations (n = 2), body
location (n = 7) and number of tags (1 or 2), there was a lack of
unique tags compared with the number of fish in a few sites and,
consequently, some individuals were only fin-clipped. Lastly, individuals were placed in well-oxygenated water and released back
into the capture site after their full-recovery. A total of 249 brown
trout were captured, among which 195 were individually tagged.
Five months later (from 12th to 14th September 2011), recaptures were carried out using the same sampling method and a
total of 446 brown trout were captured. After anaesthesia, all
individuals were checked for evidence of recapture. Recaptured
individuals
were
measured
(FLR 1 mm),
weighed
(WR 01 g), fin-clipped, and stomach contents were collected.
The same procedure was applied to unrecaptured individuals,
and scales were also collected for age determination. Young-ofthe-year juveniles that hatched in April 2011–May 2011 (i.e. during the monitoring) were rarely present and were consequently
not considered in subsequent analyses. For individuals that were
recaptured fin-clipped in September but not VIE-tagged in April,
‘individual genetic tagging’ was carried out using subsamples
of fin preserved in 70% ethanol and nine microsatellite loci
(see methodological details in Andreou et al. 2012). After DNA
extraction, PCR amplifications were run and allelic sizes
(two allele per loci) were scored using GENEMAPPER 4.0 (Applied
Biosystems, Foster City, CA, USA). Recaptured individuals were
subsequently reassigned to capture individuals using individual
© 2014 The Authors. Journal of Animal Ecology © 2014 British Ecological Society, Journal of Animal Ecology, 83, 1025–1034
1028 C. Evangelista et al.
genotypes and FLC (Andreou et al. 2012). Among the 446 brown
trout captured, 93 were tagged in spring, leading to a 477%
recapture rate. Although salmonids could be relatively mobile
(Gowan et al. 1994), we focused our analyses at the individual
level on recaptured individuals only.
To quantify trophic specialization following Ara
ujo et al.
(2007), prey items from stomach contents were counted, identified
and assigned to trophic groups as described above. Fish diet was
found to be exclusively composed of terrestrial and aquatic invertebrates, with no evidence of cannibalism. The dietary contribution of each prey taxon to individual fish was estimated by the
product of prey number by the average oven-dried (60 °C for
48 h) mass determined on 2–40 intact prey weighed to the nearest
01 mg (e.g. Ara
ujo et al. 2007). Stable isotope analyses were performed on fin samples of all recaptured individuals for each sampling period. When possible, additional fish collected in
September were analysed to reach a minimum of 15 individuals
per site. Therefore, stable isotope analyses were performed on all
recaptured individuals (93 individuals with two samples per individual), and 111 additional individuals only sampled in September (Appendix S2, Supporting information). Additionally,
potential prey were collected for stable isotope analyses in each
studied site during fish sampling (April and September). Prey
were collected according to their abundance, and samples for stable isotope analyses were composed of pooled items to account
for potential spatial variability within site (n = 3–30 individuals
per sample; e.g. Cucherousset et al. 2011). Prey were categorized
into five trophic groups: aquatic herbivores, aquatic predators,
aquatic detritivores, terrestrial herbivores and terrestrial predators
(further details available in Appendix S2, Supporting information). Within-group variability in stable isotope values was low
compared with between-group variability (Appendix S2, Supporting information), and therefore, these five functional and isotopic
groups were used for subsequent analyses.
quantification of trophic specialization,
trophic diversity and trophic niche
Trophic specialization was calculated using the integrative
method developed by Ara
ujo et al. (2007). This approach is
based on the dietary proportion of each prey group determined
from stomach contents, their d13C value and their dry mass to
calculate an index of individual trophic specialization, IS(exp),
using the relationship between expected isotopic variance and
individual trophic specialization (details available in Ara
ujo et al.
2007). The calculations were performed using the program ‘VarIso’ with 5700 simulations (Ara
ujo et al. 2007). For the sake of
clarity, we used the ‘index of specialization’ (1IS(exp)). Values
closer to 1 indicated higher degree of trophic specialization in the
population (Bolnick et al. 2002).
Trophic diversity in each population was quantified using stable isotope values (Layman et al. 2007) with ‘Stable Isotope
Bayesian Ellipse in R’, an approach developed to cope with disparities in sample size and to incorporate uncertainty in stable
isotope values (Jackson et al. 2011). Although the number
of individuals sampled in each population was representative of
population size, this ellipse-based method focused on the core of
the isotopic niche represented by the standard ellipse area (SEA).
Population trophic diversity was estimated using SEAb, the
Bayesian estimate of the SEA that was calculated using 10 000
replicates (Jackson et al. 2011).
To determine the trophic niche at the individual level, Bayesian mixing model SIAR (Parnell et al. 2010) was used to quantify the relative dietary contribution (%) of each trophic group
of prey assimilated by consumers (Layman et al. 2012). SIAR
allows the integration of variability in prey and consumer stable
isotope values, trophic enrichment factors and other unquantifiable sources of uncertainty. As the C : N ratio of invertebrates
was high, their stable isotope values were lipid corrected (Post
et al. 2007; details available in Appendix S2, Supporting information). As there is no specific trophic enrichment factors for
the model species reported in the literature, we followed a conservative approach and used trophic enrichment factors of 10&
( 10 SD) and 33& ( 10 SD) for d13C and d15N, respectively
(e.g. Cucherousset et al. 2011). Models were run separately at
tagging using stable isotope values of prey sampled in April
(analytical precision used as a variation estimated) and at recapture using the mean and standard deviation of prey sampled in
April and September. This was performed to incorporate potential temporal variability of prey stable isotope values over the
study period and to match with consumer tissue turnover
(Layman et al. 2012).
statistical analyses
Results from a previous study (Lecerf et al. 2012) and visual
examination of our data suggested that stream ecosystem
responses to canopy openness may be nonlinear. A linear model
with a quadratic term was thus used to assess monotonic (linear
and nonlinear) and non-monotonic (hump-shaped and U-shaped)
relationships between canopy openness and the response variables
(i.e. trophic resource availability, brown trout densities, specialization and trophic diversity) determined at the site level (n = 10).
To guard against overfitting, the quadratic term was removed for
models when not significant (P > 005; Crawley 2007). Canopy
openness was square-root transformed to ensure more even dispersion of sites along the gradient. When needed, response variables were transformed to meet the assumptions of linear models.
Specifically, brown trout densities and trophic diversity were log
transformed.
Trout population densities (ind. 100 m2) in stream sites were
estimated based on data collected on the second sampling date
(September). Statistics were conducted on brown trout population densities, densities of age-1 individuals and densities of
> age-1 individuals only (i.e. ‘population density’, ‘juvenile density’ and ‘adult density’, respectively) to assess if age class with
the greatest nutritional demand would respond the most to
change in canopy cover and prey availability. Linear models were
used to test the effects of the inputs of allochthonous prey on
brown trout density and to test the potential effects of brown
trout density (i.e. ‘population density’) on trophic specialization
and trophic diversity.
To determine individual responses to canopy openness, we first
tested the existence of temporal maintenance of individual trophic specialization using a repeatability approach of the dietary
contribution of aquatic predators and terrestrial predators
obtained from individual mixing models. Terrestrial and aquatic
predators were selected because of their high contribution to
trout diet over the study period (see Results). Adjusted repeatability was quantified using the intra-class correlation coefficient
(ICC) given by linear mixed models (LMM) with individual
identity as a random factor (Nakagawa & Schielzeth 2010).
© 2014 The Authors. Journal of Animal Ecology © 2014 British Ecological Society, Journal of Animal Ecology, 83, 1025–1034
Individual variability in disturbed streams
Individuals with only one measurement (i.e. individuals tagged in
April but not recaptured and untagged individuals captured in
September) were also included in the models (Kluen & Brommer
2013), and site was used as a fixed effect (Nakagawa & Schielzeth 2010). Then, individual growth rate was used as integrative
proxy of individual performances (Cucherousset et al. 2011).
Growth rate of recaptured individuals (% month1) was calculated using: Growth rate = 100 9 (lnWR lnWC)/(number of
months). The effects of trophic niche specialization on fish
growth rate were then tested using LMM with site as a random
factor (Pinheiro et al. 2012). Fixed effects were canopy openness
modelled as linear and quadratic terms to assess between-site
differences, brown trout population density and age class (i.e.
juveniles vs. adults to account for the fact that fish growth is not
isometric) and dietary contribution of selected prey groups as
indicator of individual trophic specialization within populations.
The full model also included the interaction between age and
individual trophic niche to assess age-specific response of individual performances to specialization. The interaction was removed
from the model when it was not significant (Crawley 2007). All
statistical analyses were performed using R (R Development Core
Team 2011).
Results
habitat characteristics and prey response
to canopy openness
The 10 streams had narrow channels (mean wet channel
width: 16–39 m) with shallow water (mean water depth:
12–20 cm). Riffles were consistently the dominant habitat
type (per cent channel area: 544–944%, Appendix S1,
Supporting information). Stream sites differed in mean
(March to September 2011) and maximum (July–August)
water temperature, but these variations were not significantly related to canopy openness (R2 = 026, P = 0126
and R2 = 018, P = 0229, respectively). Importantly,
stream sites differed substantially in the availability of
aquatic (Fig. 1a) and terrestrial (Fig. 1b) invertebrate
prey. Although a general increase was observed, the relationship between canopy openness and the total biomass
of aquatic prey was not significant (R2 = 029, P = 0112).
In fact, canopy openness strongly determined the biomass
of aquatic herbivores that increased substantially and linearly with canopy openness (R2 = 073, P = 0002; Fig. 1c).
No significant relationships between canopy openness,
biomass of aquatic detritivores and aquatic predators
were observed (R2 < 001, P = 0948 and R2 = 005,
P = 0552, respectively), showing no predictable responses
of these prey categories to canopy openness. Differences
in canopy openness explained significantly the inputs of
terrestrial prey that followed a U-shaped curve with the
lowest inputs of allochthonous prey measured at the sites
with intermediate levels of canopy openness (R2 = 064,
linear term: P = 0010, quadratic term P = 0009;
Fig. 1b). This trend was determined by the inputs of
terrestrial herbivores (R2 = 055, linear term: P = 0024,
quadratic term: P = 0030) rather than terrestrial predators (R2 = 022, P = 0171; Fig. 1d).
1029
population responses to canopy openness
The population density of brown trout in late summer
ranged from 73 to 375 ind. 100 m2 and was not significantly related to canopy openness (R2 = 007, P = 0481;
Fig. 2a). Juvenile density varied over a 23-fold range, and
the proportion of juveniles in populations ranged from 4
to 94%. Juvenile density showed a U-shaped curve
response to canopy openness (R2 = 052, linear term:
P = 0079, quadratic term: P = 0049; Fig. 2a), while no
significant relationship was found for the density of adults
(R2 = 001, P = 0745). In addition, no significant relationship between population density of brown trout and
the input of allochthonous prey was observed (R2 = 021,
P = 0148; Fig. 2b), but juvenile density was positively
affected by the input of allochthonous prey (R2 = 042,
P = 0043; Fig. 2b).
Population trophic specialization varied over a 4-fold
range [from 014 (95% CI: 008–021) to 050 (95% CI:
045–069)] but was not significantly associated with canopy openness (R2 = 032, P = 0090; Fig. 2c, Table 1).
The level of ecological opportunities did not explain variation in trophic specialization as no significant relationship between the inputs of terrestrial prey and trophic
specialization at the population level was detected
(R2 = 003, P = 0642). In contrast, trophic specialization
was explained by population density (R2 = 068, linear
term: P = 0033, quadratic term: P = 0024; Table 1). Specialization slightly decreased among the eight sites with
smallest trout population (mean index = 022), and it
increased substantially in the two sites with the largest
populations (Fig. 2d).
Trophic diversity of brown trout population was highly
variable across sites [from 081 (95% CI: 050–114) to
434 &2 (95% CI: 296–572)] and was tightly related to
canopy openness in a non-monotonic manner (R2 = 072,
linear term: P = 0005, quadratic term: P = 0009;
Table 1). The hump-shaped curve indicated that trophic
diversity was the highest at intermediate levels of canopy
openness (Fig. 3a). In contrast, no significant relationship
between trophic diversity and brown trout population
density was observed (R2 < 001, P = 0962; Fig. 3b,
Table 1).
individual consequences
Mixing models predicted that terrestrial and aquatic predators were, overall, the main prey consumed by brown
trout (i.e. mean = 264% ( 006 SD) and 221%
( 005 SD), respectively). The dietary contribution of
terrestrial and aquatic predators to the diet of each recaptured individual was significantly repeatable between
spring and late summer (ICC = 049, P < 0001 and
ICC = 051, P < 0001, respectively, n = 319), indicating
that individuals were specialized. The consumption of terrestrial predators had no significant effect on individual
growth rate (P = 0225; Table 2), while a higher dietary
© 2014 The Authors. Journal of Animal Ecology © 2014 British Ecological Society, Journal of Animal Ecology, 83, 1025–1034
(c)
2500
2000
1500
1000
500
0
2
4
6
8
1600
1200
800
400
0
0
2
4
6
140
120
100
80
60
40
20
0
8
0
2
4
6
8
0
2
4
6
8
(d)
Allochthonous prey inputs
(mg dry mass.m–2.d–1)
0
(b)
Allochthonous prey inputs
(mg dry mass.m–2.d–1)
Autochthonous prey biomass
(mg dry mass.m–2)
(a)
Autochthonous prey biomass
(mg dry mass.m–2)
1030 C. Evangelista et al.
140
120
100
80
60
40
20
0
Canopy openness
(%, square-root transformed)
Canopy openness
(%, square-root transformed)
Brown trout density
(ind.100 m–2, log transformed)
Fig. 1. Relationship between canopy openness (%, square-root transformed) and (a) the biomass of autochthonous prey (mg dry mass
m2), (b) the inputs of allochthonous prey (mg dry mass m2 day1), (c) the biomass of aquatic predators, herbivores and detritivores
(solid squares, open dots and crosses, respectively, in mg dry mass m2), and (d) the inputs of terrestrial predators and herbivores (solid
square and open dots, respectively, mg dry mass m2 day1) in the 10 stream sites. Significant relationships are depicted using continuous lines.
(a)
(b)
2·0
1·6
1·2
0·8
0·4
0·0
0
2
4
6
8
0
Canopy openness
(%, square-root transformed)
(c)
Index of specialization
40
80
120
160
Allochthonous prey inputs
(mg dry mass.m–2·d–1)
(d)
0·6
0·4
0·2
0·0
0
2
4
6
Canopy openness
(%, square-root transformed)
8
0·8
1
1·2
1·4
1·6
Population trout density
(ind.100 m–2, log transformed)
Fig. 2. Relationship between (a) canopy openness (%, square-root transformed) and (b) the inputs of allochthonous prey (mg dry mass
m2 day1) and the population (solid symbols) and juvenile (open symbols) density (ind. 100 m2, log transformed) of brown trout in
the 10 stream sites. Relationship between (c) canopy openness (%, square-root transformed) and (d) the population density (ind.
100 m2, log transformed) of brown trout and the index of specialization in the 10 stream sites. Significant relationships are depicted
using continuous lines.
contribution of aquatic predators was associated with a
higher growth rate (P < 0001; Table 2). This effect was,
however, age-dependent (interaction term: P = 0002;
Table 2; Appendix S1, Fig. S1, Supporting Information),
and juvenile brown trout displayed a stronger growth
advantage than adults. Canopy openness had no significant effect on individual growth rate (P > 0866; Table 2).
Discussion
The ecological importance of trophic subsidies between
riparian forest and aquatic ecosystems has now been
widely recognized, including their role in consumer–prey
dynamics and their direct effects on consumers (e.g. Baxter, Fausch & Saunders 2005; Bartels et al. 2012). The
© 2014 The Authors. Journal of Animal Ecology © 2014 British Ecological Society, Journal of Animal Ecology, 83, 1025–1034
Individual variability in disturbed streams
Table 1. Results of the simplified regression models assessing linear and quadratic relationships between canopy openness and
brown trout population density and trophic specialization and
trophic diversity (n = 10)
Response
variables
Source of
variation
Trophic
specialization
Canopy
openness
Intercept
Canopy
openness
Canopy
openness2
Intercept
Density
Density2
Intercept
Density
Intercept
Trophic
diversity
Trophic
specialization
Trophic
diversity
d.f.
Estimate
(SE)
t
P
8
003 (002)
193
0090
8
7
015 (007)
043 (011)
217
397
0062
0005
7
005 (001)
359
0009
7
7
7
7
8
8
054
341
150
213
002
021
305
264
286
274
005
050
0018
0033
0024
0029
0962
0632
(018)
(129)
(052)
(078)
(033)
(041)
Significant P-values are displayed in bold.
summer input of terrestrial prey to streams represents an
ecological opportunity for opportunistic predators, such
as brown trout, that have high-energy requirements during warmer months of the year. This idea is supported
by a large body of literature on the trophic ecology of
salmonids (Kawaguchi & Nakano 2001; Nakano & Murakami 2001) and by the positive relationship between allochthonous prey inputs and juveniles density measured in
the present study. In addition, we provide evidence that
riparian land-use and forest management may affect the
population patterns of trophic specialization and individual performances in brown trout (Fig. 4). Changes in ecological opportunities, by directly affecting the density of
consumers, may indirectly modify the degree of individual
specialization. Individual longitudinal monitoring (Ara
ujo,
Bolnick & Layman 2011) further indicates that individual
diet variability on most of the functionally important prey
was repeatable over the growing season, having subsequent effects on individual performances (Fig. 4).
Previous studies on the effects of riparian forest cover
on trophic subsidies have mainly hinted at the linear
decline of allochthonous prey inputs with riparian forest
alteration (e.g. Kawaguchi & Nakano 2001; Baxter et al.
2004; Er}
os et al. 2012). As the present study was based
on forest streams along a gradient of canopy openness
rather than a comparison between extreme cases, we were
able to detect nonlinear changes in allochthonous prey
inputs (Fig. 4). Although counterintuitive, the rise of terrestrial invertebrate inputs along the second half of the
gradient could be explained by the growth of understorey
vegetation as canopy opening enhances ground luminosity
(a)
Trophic diversity
(‰², log transformed)
1031
(b)
0·8
0·4
0·0
–0·4
0
2
4
6
Canopy openess
(%, square-root transformed)
8
0·8
1·0
1·2
1·4
1·6
Population trout density
(ind.100 m–2, log transformed)
Fig. 3. Relationship between (a) canopy openness (%, square-root transformed) and (b) population density of brown trout (ind.
100 m2, log transformed) and trophic diversity (&2, log transformed) in the 10 studied populations. The significant relationship is
depicted using continuous line.
Table 2. Results of linear mixed models used to test for both the effects of canopy openness and individual trophic specialization
[dietary contribution of aquatic (Model 1) or terrestrial (Model 2) predators] on growth rate of recaptured individuals
Model
Source of variation
d.f.
Estimate (SE)
t
P
Model 1
Age (juveniles vs. adults)
Terrestrial predators
Canopy openness
Population density
Intercept
Age (juveniles vs. adults)
Aquatic predators
Canopy openness
Population density
Age*Aquatic predators
Intercept
1,81
1,81
1,7
1,7
1,81
1,80
1,80
1,7
1,7
1,80
1,80
736 (098)
1119 (916)
003 (057)
230 (514)
1829 (655)
934 (519)
6466 (1008)
008 (043)
149 (402)
6909 (2145)
150 (706)
752
122
005
045
279
180
358
018
037
322
021
<0001
0225
0963
0669
0007
0076
< 0001
0866
0722
0002
0832
Model 2
Significant P-values are displayed in bold.
© 2014 The Authors. Journal of Animal Ecology © 2014 British Ecological Society, Journal of Animal Ecology, 83, 1025–1034
1032 C. Evangelista et al.
RIPARIAN CANOPY OPENNESS
Low
–
Intermediate
Fig. 1b-d
Subsidies
+
High
+
Subsidies
Fig. 2b
Density
–
SpecializaƟon
+
Density
Fig. 2d
Diet
repeatability
+
SpecializaƟon
+
Individual performances
Fig. 4. Synthetic representation of the effects of changes in riparian canopy openness caused by human activities on trophic subsidies and subsequent consequences on brown trout (i.e. density,
specialization and performances).
(Basset et al. 2001), thereby increasing the abundance of
ground terrestrial invertebrates. As a consequence, a
U-shaped relationship between juvenile brown trout density and canopy openness was observed (e.g. Lecerf et al.
2012), indicating that ecological opportunities can lead to
an increase population density.
Trophic specialization and terrestrial prey inputs were
not related, suggesting that factors other than resource
diversity and availability may have prevailed in determining brown trout individual specialization in our study.
Intraspecific competition is often cited as a key determinant of individual specialization (Svanb€ack & Persson
2004; Svanb€ack & Bolnick 2007; Tinker, Bentall & Estes
2008; Svanb€ack et al. 2011) because competition favours
resource partitioning and thus increases diet variation
within a population (Ara
ujo, Bolnick & Layman 2011).
Here, the intraspecific competition hypothesis is supported
by the maximum degree of individual specialization
reached in the populations with the highest densities.
Dispersion of individuals in the isotopic niche space
depicts the degree of population trophic diversity (Layman et al. 2007; Jackson et al. 2011). It peaked at intermediate sites possibly because terrestrial prey inputs are
lowest, suggesting that brown trout predation was mainly
directed towards different aquatic prey. Moreover, aquatic
prey in these streams displayed a wide range of carbon
isotope values (Appendix S2, Supporting information),
potentially leading to an increased trophic diversity of
brown trout. Aquatic invertebrates rely on mixed carbon
sources with distinct isotopic signatures; terrestrial plant
litter being 13C-depleted, and autochthonous primary producers (algae, moss) being 13C-enriched (Hoeinghaus &
Zeug 2008). Although terrestrial plant litter is the primary
energy source to forested streams (Wallace et al. 1997), its
contribution to secondary production may decrease as instream autochthonous production increases in response to
riparian canopy opening (Kiffney, Richardson & Bull
2003; England & Rosemond 2004). Autochthonous and
allochthonous carbon sources should therefore contribute
more equitably to stream food webs at intermediate canopy cover.
Multiple samples from the same individuals using individual tagging and stable isotope analyses (e.g. Cunjak
et al. 2005; Cucherousset, Paillisson & Roussel 2013) indicated that individual diet was strongly repeatable over
several months during the growth season and increased
individual performance in our study. Optimal foraging
theory predicts that consumer selects their prey to maximize energy acquisition which, in turn, should improve
individual performances (e.g. Svanb€
ack & Bolnick 2008;
Bolnick & Ara
ujo 2011; Cucherousset et al. 2011). Here,
brown trout diet was mainly composed of aquatic and terrestrial predators that are likely to be the most valuable
prey (larger with high energetic value and/or lower capture time) compared with other prey groups. However,
this idea is not fully supported by our data as only the
consumption of aquatic predators, not terrestrial predators, had a significant positive effect on individual growth
rate, a proxy of fitness. It would be of great interest to
examine the long-term repeatability on individual specialization and its ultimate consequences on individual lifehistory traits such as reproduction and survival.
In conclusion, we found that individual variability in
trophic ecology, driven by human-induced modifications
of environmental conditions that affect food resource
availability and ecological opportunities, induces complex
consequences at both individual and population levels. As
the specialization on aquatic predators by brown trout
was found to increase individual performances, we speculate that individual trophic specialization could, in turn,
affect many ecological properties. Harmon et al. (2009)
revealed that individual variability in resource use could
affect ecosystem functioning. However, the potential
effects of individual trophic specialization at higher levels
of biological organization remain poorly explored, and we
argue that further studies should investigate the relative
importance of individual trophic specialization on ecosystem functioning.
Acknowledgements
This study was funded through the French national program ‘Biodiversite,
Gestion Forestiere, et Politiques Publiques’ (SYLECOL research project)
to AL. CE and JC are members of the laboratory EDB, part of the ‘Laboratoire d’Excellence (LABEX) entitled TULIP (ANR-10-LABX-41)’. JC
was supported by an ‘ERG Marie Curie’ grant (PERG08-GA-2010276969). We are very grateful to T.K. Pool for editing the English style, to
S. Blanchet, S. Lamothe and E. Monoury for their help in the field, and
to R. Etienne, G. Loot and C. Veyssiere for laboratory assistance. Three
anonymous reviewers provided helpful comments on earlier drafts of the
manuscript.
© 2014 The Authors. Journal of Animal Ecology © 2014 British Ecological Society, Journal of Animal Ecology, 83, 1025–1034
Individual variability in disturbed streams
References
Allendorf, F.W. & Hard, J.J. (2009) Human-induced evolution caused by
unnatural selection through harvest of wild animals. Proceedings of the
National Academy of Sciences, 106, 9987–9994.
Andreou, D., Vacquie-Garcia, J., Cucherousset, J., Blanchet, S., Gozlan,
R.E. & Loot, G. (2012) Individual genetic tagging for teleosts: an
empirical validation and a guideline for ecologists. Journal of Fish Biology, 80, 181–194.
Ara
ujo, M.S., Bolnick, D.I. & Layman, C.A. (2011) The ecological causes
of individual specialization. Ecology Letters, 14, 948–958.
Ara
ujo, M.S., Bolnick, D.I., Machado, G., Giaretta, A.A. & Reis, S.F.
(2007) Using d13C stable isotopes to quantify individual-level diet variation. Oecologia, 152, 643–654.
Bartels, P., Cucherousset, J., Steger, K., Ekl€
ov, P., Tranvik, L.J. & Hillebrand, H. (2012) Reciprocal subsidies between freshwater and terrestrial
ecosystems structure consumer resource dynamics. Ecology, 93, 1173–
1182.
Basset, Y., Charles, E., Hammonds, D.S. & Brown, V.K. (2001)
Short-term effects of canopy openness on insect herbivores in a rain forest in Guyana. Journal of Applied Ecology, 38, 1045–1058.
Baxter, C.V., Fausch, K.D. & Saunders, W.C. (2005) Tangled webs: reciprocal flows of invertebrate prey link streams and riparian zones. Freshwater Biology, 50, 201–220.
Baxter, C.V., Fausch, K.D., Murakami, M. & Chapman, P.L. (2004) Fish
invasion restructures stream and forest food webs by interrupting reciprocal prey subsidies. Ecology, 85, 2656–2663.
Baxter, C.V., Fausch, K.D., Murakami, M. & Chapman, P.L. (2007)
Invading rainbow trout usurp a terrestrial prey subsidy from
native charr and reduce their growth and abundance. Oecologia, 153,
461–470.
Bolnick, D.I. (2004) Can intraspecific competition drive disruptive selection? An experimental test in natural populations of sticklebacks. Evolution, 58, 608–618.
Bolnick, D.I. & Ara
ujo, M.S. (2011) Partitioning the relative fitness effects
of diet and trophic morphology in the threespine stickleback. Evolutionary Ecology Research, 13, 439–459.
Bolnick, D.I., Yang, L.H., Fordyce, J.A., Davis, J.M. & Svanb€ack, R.
(2002) Measuring individual-level resource specialization. Ecology, 83,
2936–2941.
Bolnick, D.I., Svanb€
ack, R., Fordyce, J.A., Yang, L.H., Davis, J.M.,
Hulsey, C.D. et al. (2003) The ecology of individuals: incidence and
implications of individual specialization. The American Naturalist, 161,
1–28.
Bolnick, D.I., Ingram, T., Stutz, W.E., Snowberg, L.K., Lau, O.L. &
Paull, J.S. (2010) Ecological release from interspecific competition leads
to decoupled changes in population and individual niche width. Proceedings of the Royal Society B: Biological Sciences, 277, 1789–1797.
Crawley, M.J. (2007) The R Book. Wiley, New York, NY.
Cucherousset, J., Paillisson, J.-M. & Roussel, J.-M. (2013) Natal departure
timing from spatially varying environments is dependent of individual
ontogenetic status. Naturwissenschaften, 100, 761–768.
Cucherousset, J., Acou, A., Blanchet, S., Britton, J.R., Beaumont, W.R.C.
& Gozlan, R.E. (2011) Fitness consequences of individual specialization
in resource use and trophic morphology in European eels. Oecologia,
167, 75–84.
Cucherousset, J., Boul^etreau, S., Azemar, F., Compin, A., Guillaume, M.
& Santoul, F. (2012) “Freshwater killer whales”: beaching behavior of
an alien fish to hunt land birds. PLoS ONE, 7, e50840.
Cunjak, R.A., Roussel, J.-M., Gray, M.A., Dietrich, J.P., Catwright, D.F.,
Munkittrick, K.R. et al. (2005) Using stable isotope analysis with telemetry or mark-recapture data to identify fish movement and foraging.
Oecologia, 144, 636–646.
Darimont, C.T., Paquet, P.C. & Reimchen, T.E. (2009) Landscape heterogeneity and marine subsidy generate extensive intrapopulation niche
diversity in a large terrestrial vertebrate. Journal of Animal Ecology, 78,
126–133.
Ekl€
ov, P. & Svanb€
ack, R. (2006) Predation risk influences adaptive morphological variation in fish populations. The American Naturalist, 167,
440–452.
England, L.E. & Rosemond, A.D. (2004) Small reductions in forest cover
weaken terrestrial-aquatic linkages in headwater streams. Freshwater
Biology, 49, 721–734.
Er}
os, T., Gustafsson, P., Greenberg, L.A. & Bergman, E. (2012) Forest-stream linkages: effects of terrestrial invertebrate input and light on
1033
diet and growth of brown trout (Salmo trutta) in a boreal forest stream.
PLoS ONE, 7, e36462.
Gowan, C., Young, M.K., Fausch, K.D. & Riley, S.C. (1994) Restricted
movement in resident stream salmonids: a paradigm lost? Canadian
Journal of Fisheries and Aquatic Sciences, 51, 2626–2637.
Harmon, L.J., Matthews, B., Des Roches, S., Chase, J.M., Shurin, J.B. &
Schluter, D. (2009) Evolutionary diversification in stickleback affects
ecosystem functioning. Nature, 458, 1167–1170.
Hoeinghaus, D.J. & Zeug, S.C. (2008) Can stable isotope ratios provide
for community-wide measures of trophic structure? Comment. Ecology,
89, 2353–2357.
Jackson, A.L., Inger, R., Parnell, A.C. & Bearhop, S. (2011) Comparing
isotopic niche widths among and within communities: SIBER-Stable
Isotope Bayesian Ellipses in R. Journal of Animal Ecology, 80, 595–602.
Kawaguchi, Y. & Nakano, S. (2001) Contribution of terrestrial invertebrates
to the annual resource budget for salmonids in forest and grassland
reaches of a headwater stream. Freshwater Biology, 46, 303–316.
Kiffney, P.M., Richardson, J.S. & Bull, J.P. (2003) Responses of periphyton and insects to experimental manipulation of riparian buffer width
along forest streams. Journal of Applied Ecology, 40, 1060–1076.
Kluen, E. & Brommer, J.E. (2013) Context-specific repeatability of personality traits in a wild bird: a reaction-norm perspective. Behavioral Ecology, 24, 650–658.
Layman, C.A., Arrington, D.A., Montañ a, C.G. & Post, D.M. (2007) Can
stable isotope ratios provide for community-wide measures of trophic
structure? Ecology, 88, 42–48.
Layman, C.A., Ara
ujo, M.S., Boucek, R., Hammerschlag-Peyer, C.M.,
Harrison, E., Jud, Z.R. et al. (2012) Applying stable isotopes to examine food-web structure: an overview of analytical tools. Biological
Reviews, 87, 545–562.
Lecerf, A., Baudoin, J.M., Besson, A., Lamothe, S. & Lagrue, C. (2012)
Is smaller necessary better? Effects of small riparian forest harvesting on stream ecosystems. Annales de Limnologie - International Journal
of Limnology, 48, 401–409.
Mellina, E. & Hinch, S.G. (2009) Influences of riparian logging and
in-stream large wood removal on pool habitat and salmonid density
and biomass: a meta-analysis. Canadian Journal of Forest Research, 39,
1280–1301.
Moore, R., Spittlehouse, D. & Story, A. (2005) Riparian microclimate and
stream temperature response to forest harvesting: a review. Journal of
the American Water Resources Association, 41, 813–834.
Nakagawa, S. & Schielzeth, H. (2010) Repeatability for Gaussian and
non-Gaussian data: a practical guide for biologists. Biological Reviews,
85, 935–956.
Nakano, S., Fausch, K.D. & Kitano, S. (1999a) Flexible niche partitioning
via a foraging mode shift: a proposed mechanism for coexistence in
stream-dwelling charrs. Journal of Animal Ecology, 68, 1079–1092.
Nakano, S., Miyasaka, H. & Kuhara, N. (1999b) Terrestrial-aquatic linkages: riparian arthropod inputs alter trophic cascades in a stream food
web. Ecology, 80, 2435–2441.
Nakano, S. & Murakami, M. (2001) Reciprocal subsidies: dynamic interdependence between terrestrial and aquatic food webs. Proceedings of
the National Academy of Sciences of the United States of America, 98,
166–170.
Parnell, A.C., Inger, R., Bearhop, S. & Jackson, A.L. (2010) Source partitioning using stable isotopes: coping with too much variation. PLoS
ONE, 5, e9672.
Pinheiro, J.C., Bates, D.M., DebRoy, S. & Sarkar, D. (2012) NLME: linear and nonlinear mixed effects models. R package version, 3.1–103.
http://www.R-project.org
Post, D.M., Layman, C.A., Arrington, D.A., Takimoto, G., Quattrochi, J.
& Monta~
na, C.G. (2007) Getting to the fat of the matter: models, methods and assumptions for dealing with lipids in stable isotope analyses.
Oecologia, 152, 179–189.
Quevedo, M., Svanb€ack, R. & Ekl€
ov, P. (2009) Intrapopulation niche partitioning in a generalist predator limits food web connectivity. Ecology,
90, 2263–2274.
R Development Core Team (2011) R: A Language and Environment for
Statistical Computing. R Foundation for Statistical Computing, Vienna,
Austria. Available from URL: http://www.R-project.org/
Schindler, D.E., Leavitt, P.R., Brock, C.S., Johnson, S.P. & Quay, P.D.
(2005) Marine-derived nutrients, commercial fisheries, and production
of salmon and lake algae in Alaska. Ecology, 86, 3225–3231.
Sk
ulason, S. & Smith, T.B. (1995) Resource polymorphisms in vertebrates.
Trends in Ecology and Evolution, 10, 366–370.
© 2014 The Authors. Journal of Animal Ecology © 2014 British Ecological Society, Journal of Animal Ecology, 83, 1025–1034
1034 C. Evangelista et al.
Svanb€
ack, R. & Bolnick, D.I. (2007) Intraspecific competition drives
increased resource use diversity within a natural population. Proceedings
of the Royal Society B: Biological Sciences, 274, 839–844.
Svanb€
ack, R. & Bolnick, D.I. (2008) Food specialization. Behavioral Ecology, 5, 839–844.
Svanb€
ack, R. & Persson, L. (2004) Individual diet specialization, niche
width and population dynamics: implications for trophic polymorphisms. Journal of Animal Ecology, 73, 973–982.
Svanb€
ack, R., Rydberg, C., Leonardsson, K. & Englund, G. (2011) Diet
specialization in a fluctuating population of Saduria entomon: a consequence of resource or forager densities? Oikos, 120, 848–854.
Tachet, H., Richoux, P., Bournaud, M. & Usseglio-Polatera, P. (2010)
Invert
ebr
es d’eau douce. CNRS, Paris, France.
Tinker, M.T., Bentall, G. & Estes, J.A. (2008) Food limitation leads to
behavioral diversification and dietary specialization in sea otters. Proceedings of the National Academy of Sciences of the United States of
America, 105, 560–565.
Wallace, J.B., Eggert, S.L., Meyer, J.L. & Webster, J.R. (1997) Multiple
trophic levels of a forest stream linked to terrestrial litter inputs. Science, 277, 102–104.
Wipfli, M.S. & Baxter, C.V. (2010) Linking ecosystems, food webs, and
fish production: subsidies in salmonid watersheds. Fisheries, 35, 373–
387.
Received 27 February 2013; accepted 29 January 2014
Handling Editor: Stuart Bearhop
Supporting Information
Additional Supporting Information may be found in the online version
of this article.
Appendix S1. Habitat characteristics and effect of specialization
on individual growth rate.
Table S1. Physical and environmental characteristics of the 10 studied
stream sites.
Fig. S1. Age-dependent effect of the dietary contribution of aquatic
predators (%) on growth rate of recaptured individuals
(% month1).
Appendix S2. Stable isotope analyses
Fig. S2. d13C and d15N values of individual fish (diamonds) and
putative prey (circles) analysed in April (left panels) and September
(right panels) in the 10 studied stream sites: MOUS (a, b), FRAI
(c, d), PESQ (e, f), LAMP (g, h), PEYR (i, j), ORBI (k, l), BRG1
(m, n), LINO (o, p), BRG2 (q, r) and BERN (s, t).
© 2014 The Authors. Journal of Animal Ecology © 2014 British Ecological Society, Journal of Animal Ecology, 83, 1025–1034